Transportation (2006) 33: 115–132 Ó Springer 2006

DOI: 10.1007/s11116-005-3046-2

Developing metro systems in the People’s Republic of China:

Policy and gaps

BECKY P. Y. LOO1,* & DENNIS Y. N. LI2

1
Department of Geography, The University of Hong Kong, Pokfulam, Hong Kong
2
Metro Solutions Limited, Hong Kong
(*Author for correspondence, E-mail: [email protected])

Key words: affordability gap, financing gap, localization, non-fare revenue, technology gap,
technology transfer

Abstract. Rapid urbanization has taken place in China since the Open Policy in 1978. In
face of the growing demand for mobility in large cities, new metro systems were developed
in large cities like Guangzhou and Shanghai. At present, there are seven cities with 10
metro lines under construction. What are the policy issues and major challenges of
developing metro systems in China? This paper systematically reviews the current situation
and highlights three major gaps that Chinese cities have to overcome in planning metro
systems. They are the technology gap, the financing gap and the affordability gap. A
discussion of these gaps in the Chinese context leads us to the conclusion that early
planning and careful studies are important in the development of metro systems in China.
Moreover, the existing official criteria (population and economic power) for approving the
building of metro systems are insufficient and should be supplemented by more vigorous
evaluation criteria.

1. Rapid urbanization in China

The People’s Republic of China is the third largest country in the world,
covering an area of 9.6 million square kilometers. In 2001, China had a pop-
ulation of 1.3 billion (CSSB 2003). Administratively, the country is divided
into 23 provinces, five autonomous regions, four centrally administered
municipalities and two special administration regions. Due to the different
policy and institutional settings in Hong Kong, Macau and Taiwan, the dis-
cussion only focuses on mainland China.
Ever since the Open Policy in 1978, Chinese cities have experienced rapid
growth, both in terms of number and size. Urbanization transforms the spa-
tial distribution of people into a more compact and orderly space, thus
enabling a more effective organization of economic activities and a more
efficient provision of services to city dwellers. In 2001, cities occupied 5.10%
of the land mass in China, but it housed nearly one-quarter (23.82%) of
116

the population (USST 2003). In 1995, there were 75 cities with a population
of over half a million (USST 1997). Among them, 43 had 0.5–1 million
population, 22 had 1–2 million population and 10 had over 2 million popu-
lation. By 2001, the number of these large cities of different sizes rose to 61,
28 and 13, respectively (USST 2003). Spatially, there are great variations in
the distribution of large cities. In 2001, the eastern region had 53 cities with
over half a million population, the central region had 39 and the western
region had only 10 (USST 2003). In other words, most of the large cities
were concentrated in the more developed eastern region.
This trend of rapid urbanization has been closely associated with the high
economic growth of mainland China after the Open Policy. The Chinese
Gross Domestic Product (GDP) has increased more than 20 times since 1978
and in 2002 stood at a per capita GDP of US$989 (CSSB 2003). As an
economy grows, the demand for personal mobility also increases (Owen
1964). The need is especially evident in cities where there are high concentra-
tions of population and economic activities. Rapid urbanization, in turn,
exerts pressure on municipal governments to provide the infrastructure
required to support both private and public transport. To satisfy the
demand for higher mobility, there are two major approaches.
The first approach is based on Western experience, which consists of high
reliance on motorization and its collateral heavy investment on the construc-
tion of road network. In the many Western societies, the rapid urban sprawl
and suburbanization have primarily been supported by the expanding
expressways (Taaffe et al. 1996; Noland & Lem 2002). This approach is also
associated with low road toll elasticities and high rates of car ownership and
usage (Loo 2003). In New York, for instance, the transit share of commuter
trips was only about 10.8% in 2001 (American Dream Coalition 2004). The
second approach is based on the experience in cities, like Hong Kong and
Singapore, which focus on the development of urban mass transit systems
and adopt a policy of suppressing the growth of private modes. Hong Kong
is a case in point where about 90% of the trips of the society were made by
public transport (HKSAR Government 2003). Similarly, the public transport
system in Singapore carried as many as 51% of all motorized trips in 2001
(Singapore Land Transport Authority 2003).
Increasingly, the second approach has become more attractive to large
cities for various reasons. Firstly, it is difficult and costly to find land in cit-
ies for the continual expansion of roads and car parks. The problem is par-
ticularly acute in cities with long development histories. Secondly, the
growth rate of roads can seldom catch up with the increase in car ownership
and car usage. There is increasing evidence to show that the building of new
expressways and roads will not only trigger off more induced traffic but also
 117

a more dispersed pattern of settlement conducive to more travel (Banister

1999; Noland & Lem 2002). Thirdly, it is environmentally more desirable
because the heavy reliance on private automobiles fuelled by gasoline and
the habit of solo driving contribute significantly to air pollution (Greene &
Wegener 1997). Fourthly, a city with a healthy reliance on public transport
can provide mobility to those who cannot drive, the under-aged, the elderly,
the poor and those incapable of driving. This contributes to the social sus-
tainability of the transport system (World Bank 1996).
The provision of public transport usually takes the form of a strong fleet
of buses and/or the use of a rail-based system (Armstrong-Wright 1986).
However, ‘‘most cities where citizens use public transport often are cities
equipped with efficient rail systems’’ (Meyer & Dauby 2002, p. 5). In this
paper, metro systems refer to heavy railway systems operating on exclusive
rights-of-way, running at high speeds and serving the urban areas of cities
(Armstong-Wright 1986; Fouracre et al. 2003). Where appropriate, informa-
tion about light rail is provided for comparison. The Railway Development
Study in Hong Kong (Hong Kong Government 1993, p. 2) makes a concise
assessment of the metro systems – ‘‘As off-road mass carriers, railways
reduce pressure on the road network, stimulate land development and cause
no air pollution. They are a reliable, comfortable and high speed transport
mode. However, they are also very expensive to build, and less flexible in
their operation than bus services and financially viable only if they serve
densely populated areas.’’ As the population size of a city increases, the use
of a rail-based system becomes more appealing because it is more efficient to
move people along busy traffic corridors with metro systems.
While metro systems have the advantages of higher efficiency and being
more environmentally friendly, the capital investment needed for the construc-
tion of an urban railway system is often enormous. While the capital cost
depends on local conditions, traffic, geology and complexity of the system,
Meyer and Dauby (2002) have come up with the following reference figures for
each kilometer of track: US$9.6–30 million for at-grade light rail and US$60–
180 million for underground metros. The magnitude of the financial cost of
metro systems can be visualized with the cost of railway construction in Hong
Kong (Table 1). One sees that the average historical construction cost ranged
from US$47.4 million per kilometer for the Modified Initial System built in the
late 1970s to US$0.3 billion per kilometer completed in this millennium. In
addition to the huge initial capital cost, the operation and maintenance cost of
metro systems can also be very high. In the long run, there is also high cost
associated with asset improvement and asset replacement.
Based on our fieldwork, we identified the two most important criteria for
approving the building of metro systems in China. They are population size
 118

Table 1. Financial cost of the building of a metro system in Hong Kong.

Line Route Completion Network characteristics Historical Historical construction

length year construction cost in US
(in km) cost in US dollars dollars per km

Modified initial system 15.6 1979/1980 12.8 km underground, 2.8 km overhead, $740 million $47.4 million
15 stations, 1 depot, 1 OCC

Tsuen Wan line 10.5 1982 7.4 km underground, 1.9 km overhead, $679 million $64.7 million
1.2 km at grade, 10 stations, 1 depot

Island line 12.5 1985/86 10.5 km underground, 2 km overhead, 14 stations $1.525 billion $0.1 billion
(inclusive of 2 interchange stations), 1 depot

East Harbour crossing 4.6 1989 3.7 km underground, 0.9 km overhead, 2 stations $333 million $72.4 million

Lantau and Airport Railway 35 1998 7.5 km underground, 9.5 km overhead, $4.5 billion $0.1 billion
18 km at grade, 1 depot, 1 new OCC
to accommodate all lines

Tseung Kwan O Extension 6+1.9 2001/2002 1.9 km underground for QBR, 6 km $2.7 billion $0.3 billion
(TKE) with Quarry Bay Relief underground for TKE, 1+5 stations, 1 depot
(QBR) project

Note: OCC stands for operations control center. 1 US Dollar=7.8 HK Dollar.

Source: MTR Corporation Limited (unpubl.).
 119

Table 2. Major cities in Mainland China.

Top 12 cities in terms of population size Top 12 cities with the largest amount of
(2001 data) actually used foreign direct investment (FDI)
(2001 Data)

1 Shanghai* (12.6 million) 1 Shanghai* (US$85.5 million)

2 Beijing* (9.9 million) 2 Beijing* (US$3949.3 million)
3 Chongqing (9.0 million) 3 Tianjin* (US$3217.1 million)
4 Wuhan (7.6 million) 4 Guangzhou* (US$2752.9 million)
5 Tianjin* (7.5 million) 5 Shenzhen (US$2590.8 million)
6 Guagnzhou* (5.8 million) 6 Huizhou (US$2304.0 million)
7 Shenyang (4.9 million) 7 Dalian (US$1281.6 million)
8 Xian (4.0 million) 8 Suzhou (US$1184.8 million)
9 Hangzhou (3.8 million) 9 Dongguan (US$1147.2 million)
10 Nanjing (3.7 million) 10 Qingdao (US$902.6 million)
11 Chengdu (3.4 million) 11 Wuxi (US$885.8 million)
12 Jinan (3.2 million) 12 Zhuhai (US$864.7 million)

Note: Cities with operational metro systems are marked with asterisks.
Source: USST (2003), pp. 37–44 and pp. 197–204.

(which governs demand) and economic power (which governs affordability).

Table 2 provides a comparison between these two criteria and the status of
metro development in some major cities in China. It can be seen that the
four major economic centers with high foreign direct investment (FDI), that
is, Shanghai, Beijing, Tianjing and Guangzhou, are already served by metro
systems. In contrast, not all leading population centers are served by metro
systems. Chongqing and Wuhan with over 7 million people are some excep-
tions. Are the two official criteria sufficient for identifying cities with poten-
tials to develop metro systems in China?

2. Building metro systems under a central planning regime

Despite a switch to stronger emphasis on market forces, China still very

much relies on central planning. This is especially the case with major invest-
ment projects like metro systems. While international experience shows the
dominance of governments at local levels in initiating, leading and funding
metro systems (Allport 1990), every urban railway project in China has to be
approved by the State Council (the executive arm of the central government)
and incorporated into the relevant 5-year plan. Over time, the role of the
central government has changed.
Shen (2000) classifies the development of urban mass transit in mainland
China into three stages. Stage one starts with the mid-1960s, culminating in
the opening of the first metro line in Beijing on October 1, 1969. At this
stage, central government leadership and considerations were paramount.
120

The 23.6 km urban railway line with 17 stations was built under the overrid-
ing consideration of Cold War politics – ‘‘focus on military considerations,
taking traffic relief as the secondary objective’’ (Loo & Leung 1997). With
the support of Soviet technology, this railway was built without Western
technological input. The line was a historical breakthrough. Yet, despite the
development of the Beijing Circle Line (16.4 km) and the Tianjin Metro
(7.4 km), there was no major progress in other cities.
Stage two commenced with the economic reforms in the 1980s. There was
an influx of foreign information on the potential benefits of metro systems.
The construction principle shifted to ‘‘focus on traffic, taking defense into
consideration.’’ Strong municipal governments, such as Guangzhou and
Shanghai, played an increasingly active role in promoting and materializing
the construction of urban transit systems in their territories. Metro systems
were perceived as part of urban modernization. At this stage, the construc-
tion work of Shanghai and Guangzhou metro systems began. Many other
Chinese cities followed suit in initiating technical feasibility studies. There
was euphoria about building metro systems in many Chinese cities. How-
ever, the lack of capital and the risks associated with property development
as a funding source soon led the State Council to put brakes on new metro
projects. In 1995, the State Council promulgated the ‘‘Notice of the State
Council about the Suspension of Approving Underground Rapid Transit in
Cities’’ (Decree No. 60 [1995]). This Decree halted numerous ambitious
plans of developing urban mass transit in China after the mid-1990s.
Stage three commenced in 1997, after a decision of the central govern-
ment to go for ‘‘self-reliance’’ or localization in metro development. Apart
from the huge capital investment required, the central government was keen
to keep foreign exchange control under the current monetary system. Ren-
minbi (RMB) is not yet a fully convertible and freely traded currency. This
has led the central government to play an active role in asking municipal
governments to re-consider proposals for building metro systems. Subse-
quently, it was decreed that no more than 30% of the project cost could be
imported using foreign exchange. Localization, which was estimated to
reduce up to 15–20% of the capital cost in building metro systems, was con-
sidered important in getting central approval (Ye 1999).
In contrast to the cautious attitude of the central government, many muni-
cipal governments in China are enthusiastic in developing urban railway pro-
jects because the local economies are poised to reap the benefits of solving
urban transport problems, improving urban environment, directing urban
development, spurring urban economic development, improving urban effi-
ciency and providing convenience to urban dwellers. Under the Open Policy,
the role of local governments has evolved from providers to regulators and
 121

facilitators in the provision of transport infrastructure (Loo 1999). In the

construction of metro systems, the role of local governments can be broadly
classified into three major areas: decision-making, financing and the provision
of favorable investment environment. Table 3 summarizes the role of local
governments in developing metro systems. In particular, the role of local gov-
ernments in project financing is crucial because metro systems are public
investment. The associated risks are high, the lead time is long and the pay-
back period is even longer. Moreover, the establishment of key policies and
legal procedures to facilitate associated activities, such as the acquisition of
land and the diversion of public utilities, is also important. In summary, there
is an increasingly clear division of labor between the central and local govern-
ments with respect to the development of metro systems.

3. Challenges for the development of metro systems in China

A broad overview of the current status of metro development in mainland

China is as follows: 1. Operational metro systems. There are only a few cit-
ies with metro systems. They are Beijing (94.2 km), Guangzhou (36.8 km),
Shanghai (64.9 km), Shenzhen (21.9 km) and Tianjin (7.4 km). In addition,
Changchun (14 km) and Dalian (46.7 km) are currently served by light rail.
2. Metro systems under construction. There are seven cities with 10 lines
under construction (288 km, 98 billion RMB). Apart from the continual
expansion of the metro systems in Beijing, Shanghai, Guangzhou and Tian-
jin, two cities will have operational metro systems soon. They are Hangzhou
and Nanjing. 3. Metro systems under study. There are more than 30 cities
doing some form of preparatory work on urban mass transit. It is expected
that between 2005 and 2010, at least 1000 km of urban mass transit will be
built. Some of the more imminent projects include: Chengdu, Harbin, Sheny-
ang, Suzhou and Wuhan (authors’ fieldwork 2003).
Despite the recent remarkable achievements, the development of metro
systems in China still faces a lot of problems. Given the huge population, it
is difficult for China to emulate the ‘‘standard American dream of a house
and a car.’’ Space constraint, cost constraint and environmental consider-
ations all point to an alternative approach to discourage private transport by
the provision of a well-developed public transport network consisting of
metro systems and buses, and supplemented by paratransit. The experiences
in Hong Kong and Singapore are of useful reference.
To develop a metro system, very early planning and investment must be
made. In particular, the lead time for a railway from planning to revenue
operation is typically long (Vuchic 1981). In a study of 21 metro systems in
 122

Table 3. The role of local governments in developing metro systems.

Decision-making Project financing Provision of a favorable investment environment

Establish a steering committee Establish the project financing mechanism Establish key policies and legal procedures so as
and the input of seed money to facilitate associated activities such as land acquisition

Review and approve system Provide collaterals and guarantees Integrate railway network planning with urban
network planning reports to local financiers planning and create railway reserve areas

Determine the guiding thoughts, holistic Consider the ability to seek foreign loans Re-assess the land use pattern associated with the
principles and construction standards (as urban mass transit systems do not generate development of the urban railway network
foreign exchange from its operation)

Determine the organization Provide non-monetary incentives such as Implement the resettlement plan for city
for construction management land use, tax exemption, property dwellers affected by the project
development and revenue guarantees
Approve owner organization, and Divert public utilities, as necessary
establish the rights and responsibilities Provide tax incentives, as necessary
of the owner(s)

Determine the management methods Develop a monitoring mechanism for the use of railway
and fare policy development funds, if they were to be set up

Source: Based on Wu and Liu (2002).

 123

different parts of the world, Allport (1990) found that the average construc-
tion time per line was over 7 years. Moreover, about half of the systems had
made important changes to the original plan after construction began and
three-quarters of the projects were late in completion (Allport 1990). With-
out exception, railway projects are capital intensive by nature. Metro pro-
jects call for multi-disciplinary inputs and a highly organized management to
undertake. The railway infrastructure alone typically accounts for 30–70%
of the overall costs of metro systems (Schuchmann 2002). One of the great-
est savings of the building of metro systems is a reduction in commuting
time for the society as a whole. The time saved could be used for training
and development, thus greatly enhancing the quality of the labor force and
resulting in an accumulation of human capital (Jiang & Zhang 2002).
Municipal governments have to be far-sighted and committed in the
development of metro systems. As an economy grows, the increasing afflu-
ence and the proliferation of economic activities tend to encourage private
car ownership and usage. Since the Open Policy, China has experienced
rapid economic growth. From an urban development perspective, this has
resulted in a growing number of cities, an increase in city population (mainly
due to migration) and a rise of personal income for city dwellers. At present,
it is still too expensive for most city dwellers in mainland China to own a
car. In 2002, the per capita GDP of China was about US$989 (CSSB 2003).
In the same year, the price of an average Chinese-manufactured private car
ranged from about US$10,874 to US$27,789 (State Information Center
2003). Yet private car ownership and usage will become increasingly com-
mon as general affluence reaches a higher level.
China is thus in a dilemma. On the one hand, an early decision and
advanced planning are required to ensure that the backbone of an efficient
public transport network is in place before city dwellers find public transport
inconvenient and uncomfortable and a private car more affordable. On the
other hand, the funding source must be secured at least 6–7 years before any
fare revenue can be collected. It is only when societal benefit is put in perspec-
tive that the right decisions will be made. In this paper, the focus is put on the
three major gaps that China has to face in its development of metro systems.

3.1. Technology gap

Technology refers to both the hardware and software associated with a

metro system. The hardware consists of both the civil engineering, electrical
and mechanical engineering and electronics engineering. The software con-
sists of project management, contract management, quality management,
environmental management and customer services. By international stan-
124

dards, China’s civil engineering and basic electrical and mechanical engineer-
ing expertise is adequate. Yet, in terms of control engineering and ‘‘soft
skills,’’ mainland China still lags behind most developed countries. In fact,
China’s experience in modern control engineering and ‘‘soft skills’’ have only
built up recently with the development of the metro systems in Shanghai and
Guangzhou. This necessitates the import of such technology (at least in the
short to medium term), resulting in a drain on foreign exchange. Under the
current foreign exchange control, the technology gap cannot be bridged by
the high income and internal demand within Chinese cities alone. Unless
RMB becomes a fully convertible currency, foreign-exchange borrowing still
needs to be planned or approved by the central government. In fact, the
drain on foreign exchange was a major reason for the promulgation of
Decree No. 60 [1995] suspending the approval of metro systems in the mid-
1990s.
In mainland China, there were 33 cities having 2 million inhabitants or
above in 2002. Ten of these cities were located in Central and West China,
where foreign investment and export growth were lagging behind (Loo 2004).
Thus, it is not possible for all large cities in China to overcome the technol-
ogy gap by importing foreign technology. There is a need for Chinese cities to
develop different strategies in bridging the technology gap. Moreover, it is
important for cities at different levels to go for the appropriate standards and
technology in the choice of mass urban transit systems. For instance, smaller
cities may aim at the development of lower speed and capacity urban mass
transit systems by mainly relying on local technology. Bus lanes and other
bus priority schemes should be implemented before an attempt is made to
build a metro system. Larger cities, such as major population, economic and
political centers, may have to rely more heavily on metro systems with foreign
technology. Furthermore, it is important to recognize that alternative technol-
ogies, such as light rail transit for less heavily loaded urban lines, are avail-
able to fill the gap between the metro and the bus.
In a study by Allport (1990), it was suggested that metro systems were
only likely to be economically viable in cities of 5 million inhabitants or
more unless the city was highly linear. Moreover, Babalik-Sutcliffe (2002)
also found urban form to be an important factor in determining whether
urban rail systems are successful or not. If the population yardstick is adop-
ted as the sole criterion, there are six Chinese cities where a metro system
can be economically viable. They are Beijing, Chongqing, Guangzhou,
Shanghai, Tianjin, and Wuhan (Table 2). Only Beijing, Guangzhou, Shang-
hai and Tianjin are served by existing metro systems. Given the technology
gap, it is not surprising to find that these cities are among the top 12 cities
with the highest FDI in mainland China (Table 2). By contrast, Chongqing
 125

and Wuhan, having much lower FDI, are not currently served by metro sys-
tems. Nonetheless, some recent breakthroughs have been achieved. In Chon-
gqing, a monorail system was opened in 2004. In Wuhan, the building of the
metro system is also on the agenda of the government.
As China further opens up, a more liberal approach to the use of foreign
aid and resources to fund metro construction and operation should be con-
sidered. At present, the emphasis should perhaps still be put on localization
and standardization to ensure economies of scale in the construction of
metro systems in China. In the longer run, both local and foreign technology
(in terms of hardware as well as software) should be taped to ensure that the
metro systems are not only cost effective but also reliable, safe, and efficient.
Generally, there needs to be a stronger emphasis on the adoption of modern
‘‘soft skills’’ to maximize the efficiency and effectiveness of the sophisticated
railway technology.

3.2. Financing gap

The financing gap goes beyond the ability to raise the initial capital cost of
developing the metro system. More importantly, it refers to the ability of the
metro system to generate sufficient revenue to make the system sustainable,
that is, to recover the capital investment in the medium run and to pay for
the cost of asset improvement and replacement in the long run. In China,
the Build, Operate and Own (BOO) and Build, Operate and Transfer (BOT)
models have been widely adopted in the construction of transport infrastruc-
ture, such as container ports and expressways (Loo 1999). Moreover, the
issuing of bonds and shares has also been actively considered by municipal
governments to pay for the building of metro systems (Jin & Wang 2002). In
the longer run, the development of a sinking fund for railway asset renewal
may also be necessary to meet the future cost of asset improvement and
replacement. Given the high financial cost, most metro systems worldwide
are operating at a loss and are supported by government subsidies (Allport
1990; Schuchmann 2002; Fouracre et al. 2003). Nonetheless, the transit sub-
sidy policy would induce local governments ‘‘to select capital-intensive
investment options that may not be efficient or effective’’ (Li & Wachs 2004,
p. 43).
What are the possible ways out? Without government intervention,
urban mass transit investment was found to bring about an increase in the
land value along the transit line, especially around the key metro stations
(Boyce et al. 1976). Income generated from the railway-related property
development and its associated activities can supplement fare revenue in
making metro systems financially more viable. The MTR Corporation
126

Limited in Hong Kong is a good case in point. In 2002, the profit on prop-
erty developments alone reached $3755 million. In addition, the rental and
management income arising from its property development contributed
another $987 million. Such income greatly supplemented the company’s fare
revenue ($5720 million) in making the MTR one of the most financially
sound metro systems in the world (MTR Corporation Limited 2003).
An issue closely related to the financing gap is the need for institutional
support and integrated land-use and transport planning. A transport policy
in favor of metro ridership and an integrated planning between land-use and
railway are crucial to the economic viability of the system because the den-
sity of development and the level of inter-modal competition along the rail-
way line will directly affect the potential passenger volume and financial
revenue of the system. In other words, ‘‘mere population figures are not
enough: settlement density patterns are relevant since the high investment
costs are only justified in heavily travelled corridors’’ (Meyer & Dauby 2002,
p. 6). In fact, the high-rise, high-density and linear form of urban develop-
ment in Hong Kong played a key role in contributing to the high patronage
of the metro system in Hong Kong (Tong & Wong 1997). Moreover, the
Hong Kong government adopts a rail-based transport policy and a multi-
faceted pricing strategy to discourage private car ownership and usage (Loo
2003). Integrated planning is particularly important in China because many
Chinese cities have a rather dispersed development pattern with low-rise
buildings in most parts of the city.
Table 4 shows a comparison of the population densities in different major
Chinese cities. The case of Hong Kong is also shown to allow a comparison.
In terms of city population density, Hong Kong is way above the top 12 cit-
ies in mainland China. In 2001, the average population density was 6125 per-
sons per square kilometer in Hong Kong. While in mainland China, the
highest densities were only 2409 and 2382 persons in Chengdu and Shanghai,
respectively. Yet, the population density can be much higher if the size of the
developed areas within a city is small. In Chinese terminology, developed
land areas refer to the areas with connection to the urban core and with the
provision of public utilities. In other words, developed land areas are already
built-up areas in urban or suburban areas, with more urban land uses, infra-
structure and public utilities (Yao & Shuai 1995). Most mainland Chinese cit-
ies have relatively low percentages of developed areas, ranging from 2% in
Chongqing and Wuhan to 16% in Chengdu. The percentage in Hong Kong
was about 17%. By considering the size of the developed areas, the prospect
of developing metro systems in mainland China seems brighter. While the
population density in Hong Kong was still the highest (36,675 persons per
square kilometer), the population densities of Wuhan (35,766 persons per
 127

Table 4. A comparison of city population densities among large cities.

City population/city City area Percentage of City population /

area (persons per km2) (km2) developed area developed area
(%) (persons per km2)

Shanghai 2382 5299 10 22,953

Beijing 588 12,574 6 12,668
Chongqing 607 14,875 2 33,697
Wuhan 893 8494 2 35,766
Tianjin 1008 7418 6 17,641
Guangzhou 1551 3719 14 10,969
Shenyang 1396 3495 7 20,491
Xian 2037 1964 10 21,395
Hangzhou 1237 3068 7 16,718
Nanjing 1431 2599 8 17,542
Chengdu 2409 1418 16 14,979
Jinan 914 3527 5 18,857
Hong Kong 6125 1098 17 36,675

Note: Mainland cities are arranged in descending order by population size in 2001 (see Table 2).
Source: Loo (2003); USST (2003), pp. 101–108, pp. 475–478, pp. 571–574; CSD (www page,
accessed 2004).

square kilometer) and Chongqing (33,697 persons per square kilometer) came
very close. In these cities, highly traveled corridors may be identified for the
consideration of the building of metro systems.

3.3. Affordability gap

Another gap that needs to be addressed is the affordability gap. When a

decision to build a metro system is made, the local government has to con-
sider many factors. In particular, the metro system has to be competitive in
price. At present, most of the public transit modes (namely trams and buses)
in China are operated and/or subsidized by local governments. As a result,
their fares are arbitrarily low. In most large cities, bus fare (for non-air-
conditioned buses) within the urban areas is only US$0.1 per trip. When a
metro system operates, the fare charged cannot be out of line with the other
public transit. Moreover, it is important to ensure that the low-income
groups are not excluded because of the high metro fares. In Guangzhou, for
example, the distance-based metro fare ranged from US$0.2 to US$1.0
(Guangzhou Metro Corporation 2004). In Shanghai, a similar distance-based
metro fare of US$0.2–US$0.7 per trip applies (Shanghai Metro Operation
Company Limited 2004). At present, the Beijing metro system still charges a
flat rate of US$0.4 per trip but any inter-change of the new Batung Line
with Lines 1 and 2 involves a surcharge of US$0.2. Recently, there have
128

been lively discussions about changing the fare structure to a distance-based

one, but affordability was a critical issue (http://www.xinhuanet.com 2003).
Table 5 shows the current GDP and GDP per capita of the top 12 cities
with the largest population. It can be seen that the economic power was the
highest in Shanghai (US$59.1 billion), Beijing (US$32.6 billion), Guangzhou
(US$29.6 billion), Tianjin (US$19.9 billion) and Wuhan (US$16.3 billion)
followed. In all these cases, their GDP was still very much lower than that
of Hong Kong (US$162.8 billion).
Allport (1990) estimated that the income per head should reach above
US$1800 per head for a metro system to be economically viable. Obviously,
this yardstick is rather crude and needs to be followed by a detailed cost-
benefit analysis (Armstrong-Wright 1986). Nonetheless, it serves as a general
indicator of the economic well-being of the citizens for them to afford a
metro system. Should the 2002 GDP per capita at current prices be consid-
ered, most of the top 12 cities in China can meet this criterion (Table 5).
Only Chongqing (US$1181 per capita) was below the yardstick. Taking into
account inflation by deflating the current GDP per capita to the 1991 fixed
price, the picture is less rosy. Only Guangzhou (with a GDP per capita of
US$2651), Shanghai (with a GDP per capita of US$2525) and Hangzhou
(with a GDP per capita of US$1968) passed the test. Beijing (with a GDP
per capita of US$1704) and Nanjing (with a GDP per capita of US$1653)
came close to this criterion.

Table 5. A comparison of economic power among large cities, 2001.

GDP GDP per capita GDP per capita

(at current price) (at current price) (at 1991 price) US$
100 million US$ US$

Shanghai 591 4928 2525

Beijing 326 3326 1704
Chongqing 106 1181 605
Tianjin 199 2667 1366
Wuhan 163 2161 1107
Shenyang 128 2626 1346
Guangzhou 296 5174 2651
Chengdu 94 2774 1421
Nanjing 119 3226 1653
Xian 77 1933 991
Hangzhou 144 3840 1968
Jinan 99 3098 1587
Hong Kong 1628 24,211 16,265

Note: Mainland cities are arranged in descending order by population size in 2001 (see Table 2).
1 US Dollar=8.3 Yuan.
Source: USST (2003), pp. 117–124; CSD (www page, accessed 2004).
 129

As incomes grow, the demand for personal mobility will rise. However,
investment decisions have to be made before the general affluence level hits
the threshold of car ownership. Metro systems are expensive to build, oper-
ate and maintain. The current income of most city dwellers in China cannot
suffice to pay a fare that reflects the true cost of capital investment, opera-
tion and maintenance as well as asset replacement. As shown in the case of
Hong Kong, non-fare revenue can be a major supplementary source of
income. However, fare revenue remained the largest and most important
source of income of the metro system (MTR Corporation Limited 2003).
Thus, once a metro fare is set, a mechanism has to be in place to adjust the
fare level to match the general inflation and average wage level of the city.
Figure 1 shows the trends in the MTR fares of Hong Kong in the last two
decades. The year 1980 is chosen as the base year to compare the changes in
the nominal payroll, consumer price index and average MTR fare. It can be
seen that the growth rate of the average MTR fare (6.6% per annum) was
roughly the same as that of the consumer price index (6.8% per annum).
Moreover, the increase was found to be acceptable to the society as the
nominal payroll actually rose more rapidly (11.7% per annum) during the
same period. A comparison of the average adult fares charged by the major
metro systems in the world shows that the average adult fare ranged from a
high of US$1.5 per trip in Munich and London to a low of US$0.2–0.3 per
trip in Athens, Rome and Lisbon (MTR Corporation Limited 1995, unpub-
lished). In Hong Kong, it was about US$0.7 per trip. These figures, together
with the current local public transit fares in mainland China, furnish some
ideas to the municipal governments in their evaluation of whether the citi-
zens can afford the metro system. Nonetheless, the evaluation should take
into account the injection of local government investment and subsidies,
should these be available.

Figure 1. Growth of average MTR fares in perspective, 1980–2000. Source: MTR Corporation
Limited (unpubl.).
130

4. Conclusions

Generally, the metro systems in developing countries are not financially viable
but they may be economically justifiable. Most of the savings from the build-
ing of a metro system will be in the form of time savings and a relief of dis-
comfort on other public transport modes. These benefits (highly associated
with the opportunity cost of the use of time) were estimated to constitute 76%
of the total economic return of metro systems (Allport 1990). In China, the
technology gap, the financing gap and the affordability gap are three major
issues that local governments have to consider very carefully before a metro
system is planned and built. As shown in the above discussion, the three gaps
are closely interrelated. Moreover, unless RMB becomes a floating currency,
there is a need for some forms of central planning and coordination in the
building of metro systems among Chinese cities. In particular, a coordinated
nation-wide plan would slow down the current ‘‘race’’ among municipal gov-
ernments to conduct feasibility studies and to get central approval for their
metro plans. To the central government, the two criteria of population size
and economic power for developing metro systems should be supplemented by
more vigorous evaluation exercises to ensure that the metro systems having
the highest social benefits can proceed with the necessary central and local
commitments required for raising capital and subsequent operations. In sum-
mary, the construction of a metro system is likely to be one of the largest
transport infrastructural projects of a city. Hence, much political commitment
is required and careful considerations must be made about the long-term
impact of the metro systems on the local governments and the society at large.

Acknowledgements

This project is financially supported by the Conference and Research Com-

mittee Grant (10203812). The authors would like to express their gratitude
to the hospitality of the relevant authorities in the fieldwork to China during
November, 2003.

References

Allport RJ (1990) The metro: Determining its viability. In: Margaret JH (ed) Developing World
Transport (pp. 64–69). London: Grosvenor Press International.
American Dream Coalition (2004) Transit Data. See http://www.americandreamcoalition.org/
transitdata.html.
Armstrong-Wright A (1986) Urban Transit Systems – Guidelines for Examing Options. Washington,
D.C.: World Bank.
Babalik-Sutcliffe E (2002) Urban rail systems: Analysis of the factors behind success. Transport
Reviews 22(4): 415–447.
 131

Banister D (1999) Planning more to travel less. Town Planning Review 70(3): 313–338.
Boyce DE Allen WB & Tang F (1976) Impact of rapid transit on residential-property sales prices.
In: Chatterji M (eds) Space Location and Regional Development (pp. 145–153). London: Pion.
China State Statistics Bureau (CSSB) (2003) China Statistical Yearbook 2003. Beijing: China
Statistics Press.
Fouracre F Dunkerley C & Gardner G (2003) Mass rapid transit systems for cities in the devel-
oping world. Transport Review 23(3): 299–310.
Greene DL & Wegener M (1997) Sustainable transport. Journal of Transport Geography 5(3):
117–190.
Guangzhou Metro Corporation (2004) Fare Information. See http://www.gzmtr.com/serve/guide/
ticket/value.htm.
Hong Kong Government (1993) The Railway Development Study. Hong Kong: Transport Branch,
Government Secretariat.
Hong Kong Special Administration Region (HKSAR) Government (2003) Hong Kong: The Facts –
Transport. See http://www.info.gov.hk/hkfacts/transpt.pdf.
Jiang ZZ & Zhang WJ (2002) Urban mass transit and urban economic development. Urban Mass
Transit 4: 17–21 (in Chinese).
Jin EW & Wang XM (2002) Financing and investment in metro construction against the present
financial situation. Urban Mass Transit 3: 13–16 (in Chinese).
Li JL & Wachs M (2004) The effects of federal transit subsidy policy on investment decisions: The
case of San Francisco’s Geary Corridor. Transportation 31: 43–67.
Loo BPY (1999) Development of a regional transport infrastructure: Some lessons from the
Zhujiang Delta, Guangdong, China. Journal of Transport Geography 7: 43–63.
Loo BPY (2003) Tunnel traffic and toll elasticities in Hong Kong: Some recent evidence for
international comparisons. Environment and Planning A 35: 249–276.
Loo BPY (2004) Export expansion in the People’s Republic of China since 1978: A case study of
the Pearl River Delta. China Quarterly 177: 133–154.
Loo BPY & Leung CK (1997) Characteristics of transport infrastructural development in the Pearl
River Delta under the Open Policy. Acta Geographica Sinica 52(Supplement): 28–38.
MTR Corporation Limited (2003) Annual Report 2002. Hong Kong: MTR Corporation.
Meyer W & Dauby L (2002) Why is rail transport so attractive? The general situation of suburban
rail transport throughout the world. Public Transport International (October 2002): 4–7.
Noland RB & Lem LL (2002) A review of the evidence for induced travel and changes in trans-
portation and environmental policy in the US and the UK. Transportation Research Part D 7:
1–26.
Owen W (1964) Strategy for Mobility: Transportation for the Developing Countries. Honolulu: East-
West, Center Press.
Schuchmann A (2002) The importance and difficult nature of rail infrastructure for metro systems.
Public Transport International 10: 8–11.
Shanghai Metro Operation Company Limited (2004) Fare Information. See http://www.shmetro.
com/kyfw/pjcx.asp.
Shen JY (2000) Develop urban railway transit at different levels. Urban Mass Transit 1: 11–13
(in Chinese).
Singapore Land Transport Authority (2003) News Release – Opening of Bus Lanes to Private Buses.
See http://app.lta.gov.sg/corp_press_content.asp?start=365.
State Information Center (2003) China’s Automobile Market 2003. Beijing: Machinery Industry
Press.
Taaffe EJ Gauthier HL & O’Kelly ME (1996) Geography of Transportation. New Jersey: Prentice
Hall.
Tong CO & Wong SC (1997) The advantages of a high density, mixed land use linear urban
development. Transportation 24: 295–307.
132

Urban Socio-economic Survey Team of China State Statistics Bureau (USST) (1997) Urban
Statistical Yearbook of China 1996. Beijing: China Statistics Press.
Urban Socio-economic Survey Team of China State Statistics Bureau (USST) (2003) Urban
Statistical Yearbook of China 2002. Beijing: China Statistics Press.
Vuchic VR (1981) Urban Public Transportation – Systems and Technology. Englewood Cliffs, New
Jersey: Prentice Hall.
World Bank (1996) Sustainable Transportation: Priorities for Policy Reform. Washington, D.C.:
World Bank.
www.xinhuanet.com (2003) News on 13 December 2003. See http://news.xinhuanet.com/newscen-
ter/2003-12/13/content_1228982.htm.
Wu Y & Liu SN (2002) Roles of the local government in the construction of UMT. Urban Mass
Transit 4: 22–24 (in Chinese).
Yao S & Shuai J (1995) City Land Use and the Growth of City – Example of City Expansion in
Southeast China. Hefei: Zhongguo kexue jishu daxue chubanshe (in Chinese).
Ye DQ (1999) An effective measure to reduce the cost of metro construction. Urban Mass Transit 1:
4–6 (in Chinese).

About the authors

Dr. Becky P.Y. Loo is an Associate Professor in the Department of Geography at the
University of Hong Kong. She joined the Department as a full-time Lecturer in 1997.
Her major research interests are transport, spatial analysis, economic geography and
China Studies.

Dennis Y.N. Li is the Principal Consultant of MetroSolutions Limited, Hong Kong.

He has led major metro consultancy studies in different parts of Asia, including Tai-
wan and mainland China.